The Myth of Consensus at Scale

In a startup of five people, consensus is natural. You sit in a room, you argue, you agree, you build.

In an organization of fifty, consensus is a bottleneck. In an organization of five hundred, consensus is paralysis.

The Comfort of Agreement

We seek consensus because it diffuses responsibility. If we all agreed on the decision, then no single person is to blame if it fails. It is a social safety mechanism.

But product development is not a democracy. It is a series of bets made on incomplete information. Requiring everyone to agree on the bet means you will only ever make the safest, most average wagers. You will regress to the mean.

Disagree and Commit

The alternative is not dictatorship, but clear ownership.

The role of a leader is not to make everyone happy with the decision. It is to ensure the decision is made with the best available context, and then to drive alignment around execution.

Amazon’s principle of “Have Backbone; Disagree and Commit” is often quoted but rarely practiced well. The “Commit” part is easy to say but hard to do. It requires emotional maturity. It means that once the decision is made, you must support it as if it were your own. You cannot wait in the wings for it to fail so you can say, “I told you so.”

Velocity over Perfection

Most decisions are reversible. Jeff Bezos calls these “Type 2” decisions. They are like walking through a door—if you don’t like what’s on the other side, you can walk back.

Consensus treats every decision like a “Type 1” decision—a one-way door. We debate the color of a button or the naming of a variable for days, acting as if the wrong choice will doom the company.

It won’t.

The only thing that will doom the company is moving too slowly to learn. Speed is the ultimate competitive advantage. And speed requires the ability to make a call, be wrong, fix it, and move on, without holding a committee meeting to approve the correction.